R, RStudio ve RMarkdown ile Tekrarlanabilir Rapor¹

1 Tekrarlanabilir Analiz ve Rapor

---

1.1 Replication Crisis

https://en.wikipedia.org/wiki/Replication_crisis

---

1.2 Replication Crisis Excel Version

---

2 RStudio ile proje oluştur

---

3 R Notebook

3.1 R Notebook dökümanı oluşturma

---

3.2 R Notebook’tan html, pdf ve word oluşturma

---

3.3 RNotebook vs RMarkdown

https://youtu.be/zNzZ1PfUDNk

---

4 R Markdown

4.1 Hem kendi kodları hem de html kodları yazılabilir

https://rmarkdown.rstudio.com

What is R Markdown? from RStudio, Inc. on Vimeo.

---

4.2 R Markdown: The Definitive Guide

https://bookdown.org/yihui/rmarkdown/

---

4.3 R Markdown syntax

https://gist.github.com/MinhasKamal/7fdebb7c424d23149140

---

4.4 Remedy Package

---

4.4.1 Remedy

---

4.5 R Markdown paket ve şablonları

https://bookdown.org/yihui/rmarkdown/document-templates.html

---

4.6 Render Markdown via code

inside R

markdown::markdownToHTML('markdown_example.md', 
'markdown_example.html')

command line

R -e "markdown::markdownToHTML('markdown_example.md',
'markdown_example.html')"

---

4.7 pandoc Rstudio integration

command line

export PATH=$PATH:/Applications/RStudio.app/Contents/MacOS/pandoc

R -e "rmarkdown::render('markdown_example.md')"

---

5 RMarkdown chunk içinde R kodlarını çalıştırma

{r, results='asis'}
 iris %>%
  tibble::as_tibble() %>%
  details::details(summary = 'tibble')

---

6 Metin arasında R kodlarını çalıştırma

---

7 Chunk Options

7.1 Global Options

{r global_options, include=FALSE}
knitr::opts_chunk$set(fig.width = 12,
                      fig.height = 8,
                      fig.path = 'Figs/',
                      echo = FALSE,
                      warning = FALSE,
                      message = FALSE,
                      error = FALSE,
                      eval = TRUE,
                      tidy = TRUE,
                      comment = NA)

---

7.2 Other Code Languages

---

8 R Markdown kod örneği

{r}
data("cancer")
cancer
foreign::write.foreign(df = cancer,
                        datafile = "data/cancer.sav",
                        codefile = "data/cancer.spo",
                        package = "SPSS"
                        )

---

9 R Markdown Paket Çağırma 📦

{r}
suppressPackageStartupMessages(library("tidyverse"))
suppressPackageStartupMessages(library("survival"))

9.1 Sık kullandığım paketler 📦

{tidyverse} {tidylog}

{lubridate} {janitor}

{readxl} {foreign}

{summarytools} {ggstatsplot} {tangram} {finalfit} {psycho} {jmv}

{survival} {survminer}

{report} {kableExtra}

---

10 R Markdown Veri Yükleme SPSS

---

11 R Markdown Veri Yükleme Excel

---

12 Veri Düzenleme

{r, eval=FALSE, include=FALSE, eval=FALSE, include=FALSE}
mydata <- janitor::clean_names(mydata)
cat(names(mydata), sep = ",\n")

{r}
glimpse(mydata)

{r, eval=FALSE, include=FALSE, eval=FALSE, include=FALSE}
## Recoding mydata$cinsiyet into mydata$Cinsiyet
mydata$Cinsiyet <- recode(mydata$cinsiyet,
               "K" = "Kadin",
               "E" = "Erkek")
mydata$Cinsiyet <- factor(mydata$Cinsiyet)

{r recode TNM stage, eval=FALSE, include=FALSE}
mydata$Tstage <- stringr::str_match(mydata$patolojik_evre, paste('(.+)', "N", sep=''))[,2]

mydata$Nstage <- paste0("N",
    stringr::str_match(mydata$patolojik_evre, paste( "N", '(.+)', "M", sep=''))[,2]
    )

mydata$Mstage <- paste0("M", 
    stringr::str_match(mydata$patolojik_evre, paste("M", '(.+)', sep=''))[,2]
)

{r recode TNM2, eval=FALSE, include=FALSE}
mydata <- mydata %>% 
    mutate(
        T_stage = case_when(
            grepl(pattern = "T1", x = .$Tstage) == TRUE ~ "T1",
            grepl(pattern = "T2", x = .$Tstage) == TRUE ~ "T2",
            grepl(pattern = "T3", x = .$Tstage) == TRUE ~ "T3",
            grepl(pattern = "T4", x = .$Tstage) == TRUE ~ "T4",
            TRUE ~ "Tx"
        )
    ) %>% 
mutate(
        N_stage = case_when(
            grepl(pattern = "N0", x = .$Nstage) == TRUE ~ "N0",
            grepl(pattern = "N1", x = .$Nstage) == TRUE ~ "N1",
            grepl(pattern = "N2", x = .$Nstage) == TRUE ~ "N2",
            grepl(pattern = "N3", x = .$Nstage) == TRUE ~ "N3",
            TRUE ~ "Nx"
        )
    ) %>% 
mutate(
        M_stage = case_when(
            grepl(pattern = "M0", x = .$Mstage) == TRUE ~ "M0",
            grepl(pattern = "M1", x = .$Mstage) == TRUE ~ "M1",
            TRUE ~ "Mx"
        )
    )

{r, eval=FALSE, include=FALSE}
mydata <- mydata %>% 
    mutate(
TumorPDL1gr1 = case_when(
        t_pdl1 < 1 ~ "kucuk1",
        t_pdl1 >= 1 ~ "buyukesit1"
    )
    ) %>% 
mutate(
TumorPDL1gr5 = case_when(
        t_pdl1 < 5 ~ "kucuk5",
        t_pdl1 >= 5 ~ "buyukesit5"
    )
    )   %>% 
mutate(
inflPDL1gr1 = case_when(
        i_pdl1 < 1 ~ "kucuk1",
        i_pdl1 >= 1 ~ "buyukesit1"
    )
    ) %>% 
mutate(
inflPDL1gr5 = case_when(
        i_pdl1 < 5 ~ "kucuk5",
        i_pdl1 >= 5 ~ "buyukesit5"
    )
    )

{r, eval=FALSE, include=FALSE}
mydata$sontarih <- janitor::excel_numeric_to_date(as.numeric(mydata$olum_tarihi))

{r, eval=FALSE, include=FALSE}
mydata$Outcome <- "Dead"
mydata$Outcome[mydata$olum_tarihi == "yok"] <- "Alive"

---

13 R Markdown Tanımlayıcı İstatistikler

{r}
library(summarytools)
view(dfSummary(colon_s))

---

13.1 Table One

{r, results='asis'}
# cat(names(mydata), sep = " + \n")
library(arsenal)
tab1 <- tableby(~ Cinsiyet + 
Yas + 
TumorYerlesimi
                ,
                data = mydata)
summary(tab1)

13.2 The Grammar of Tables

tangram: The Grammar of Tables

A grammar of tables

Grammar of Tables?

Easily generate information-rich, publication-quality tables from R

13.3 Kategorik Veriler

{r ", names(mydataCategorical)[i], "}
mydataCategorical %>% 
  janitor::tabyl(", names(mydataCategorical)[i], ") %>%
  adorn_pct_formatting(rounding = 'half up', digits = 1) %>%
  knitr::kable()

{r crosstable", dependent_variable, "}
mydata %>%
    summary_factorlist(dependent = '", dependent_variable, "', 
                       explanatory = explanatory,
                       total_col = TRUE,
                       p = TRUE,
                       add_dependent_label = TRUE) -> table

knitr::kable(table, row.names = FALSE, align = c('l', 'l', 'r', 'r', 'r'))

---

13.4 Kategorik Veriler için Grafikler

{r ggstatplot ", dependent_variable , " vs ", mydataCategorical_variable, ", layout='l-page'}
mydata %>% 
    ggstatsplot::ggbarstats(data = ., main = ", mydataCategorical_variable, ", condition = ", dependent_variable, ")

---

13.5 Continious Variables

{r ", names(mydataContinious)[i], "}
mydataContinious %>% 
jmv::descriptives(
    data = .,
    vars = ", names(mydataContinious)[i], ",
    hist = TRUE,
    dens = TRUE,
    box = TRUE,
    violin = TRUE,
    dot = TRUE,
    mode = TRUE,
    sd = TRUE,
    variance = TRUE,
    skew = TRUE,
    kurt = TRUE,
    quart = TRUE)

---

14 R Markdown Recode

recode

---

15 R Markdown örneği Çapraz Tablolar

library(finalfit)

{r crosstable", dependent_variable, ", message=FALSE, warning=FALSE}
mydata %>%
    summary_factorlist(dependent = '", dependent_variable, "', 
                       explanatory = explanatory,
                       column = TRUE,
                       total_col = TRUE,
                       p = TRUE,
                       add_dependent_label = TRUE,
                       na_include=FALSE
                       # catTest = catTestfisher
                       ) -> table

knitr::kable(table, row.names = FALSE, align = c('l', 'l', 'r', 'r', 'r'))

---

16 R Markdown örneği Sağkalım

• Drawing Survival Curves Using ggplot2
  https://rpkgs.datanovia.com/survminer/reference/ggsurvplot.html

16.1 Kaplan-Meier

{r Kaplan-Meier}
mydata %>%
  finalfit::surv_plot(dependent,
                      explanatory,
                      xlab='Time (months)',
                      pval=TRUE,
                      legend = 'none',
                      break.time.by = 12,
                      xlim = c(0,60),
                      legend.labs = c('a','b')
)

16.2 Sağkalım Tabloları

{r}
km_fit <- survfit(dependent ~ explanatory,
                  data = mydata)
km_fit

{r, eval=FALSE, include=FALSE}
library(survival)
km <- with(mydata, Surv(OverallTime, Outcome2))
# head(km,80)
# plot(km)

{r 1-3-5-yr}
summary(km_fit, times = c(12,36,60))

17 Pairwise comparison

{r eval=FALSE, include=FALSE}
survminer::pairwise_survdiff(formula = Surv(time, Outcome) ~ ShellAnteriorOnly, 
                             data = mydata,
                             p.adjust.method = "BH")

18 Multivariate Analysis Survival

{r Multivariate Analysis, eval=FALSE, include=FALSE}
library(finalfit)
library(survival)
explanatoryMultivariate <- explanatoryKM
dependentMultivariate <- dependentKM

mydata %>%
  finalfit(dependentMultivariate, explanatoryMultivariate) -> tMultivariate

knitr::kable(tMultivariate, row.names=FALSE, align=c("l", "l", "r", "r", "r", "r"))

{r define survival time, eval=FALSE, include=FALSE}
mydata$int <- lubridate::interval(
  lubridate::ymd(mydata$CerrahiTarih),
  lubridate::ymd(mydata$SonTarih)
  )
mydata$OverallTime <- lubridate::time_length(mydata$int, "month")
mydata$OverallTime <- round(mydata$OverallTime, digits = 1)

{r, eval=FALSE, include=FALSE}
mydata$OverallTime <- mydata$genel_sagkalim

{r, eval=FALSE, include=FALSE}
## Recoding mydata$Outcome into mydata$Outcome2
mydata$Outcome2 <- recode(mydata$Outcome,
               "Alive" = "0",
               "Dead" = "1")
mydata$Outcome2 <- as.numeric(mydata$Outcome2)

---

19 jamovi ve R entegrasyonu

Rj Editor – Analyse your data with R in jamovi

20 {jmv} paket kodları

jamovi syntax mode

---

21 Güncellemeler olunca kodlar çalışacak mı?

---

21.1 Paket Kütüphaneleri

• packrat / renv

https://environments.rstudio.com

---

21.2 Docker

• docker

---

21.2.1 The Rocker Project

Docker Containers for the R Environment

docker run --rm -ti rocker/r-base

Or get started with an RStudio® instance:

docker run -e PASSWORD=yourpassword --rm -p 8787:8787 rocker/rstudio

and point your browser to localhost:8787 Log in with user/password rstudio/yourpassword

---

Managing containers

---

21.3 Yeni R sürümleri

• RSwitch

https://rud.is/rswitch/

• Using RSwitch

https://rud.is/rswitch/guide/

: scale 30%

---

22 Yedeklemeyi nasıl yapacağız

22.1 Projeyi düzgün organize edin

• pdf
• R
• images
• bib

{r load library, include=FALSE}
source(file = here::here("R", "loadLibrary.R"))

---

22.2 Save Final Data

{r}
saved data after analysis to `mydata.xlsx`.

save.image(file = here::here("data", "mydata_work_space.RData"))

readr::write_rds(x = mydata, path = here::here("data", "mydata_afteranalysis.rds"))

saveRDS(object = mydata, file = here::here("data", "mydata.rds"))

writexl::write_xlsx(mydata, here::here("data", "mydata.xlsx"))

paste0(rownames(file.info(here::here("data", "mydata.xlsx"))), " : ", file.info(here::here("data", "mydata.xlsx"))$ctime)

22.3 GitHub

{r github push}
CommitMessage <- paste("updated on ", Sys.time(), sep = "")
wd <- getwd()
gitCommand <- paste("cd ", 
                    wd,
                    " \n git add . \n git commit --message '",
                    CommitMessage,
                    "' \n git push origin master \n",
                    sep = ""
                    )
system(command = gitCommand,
       intern = TRUE
)

---

22.4 GitHub Yedekleme

CommitMessage <- paste("updated on ", Sys.time(), sep = "")
wd <- getwd()
gitCommand <- paste("cd ", 
                    wd,
                    " \n git add . \n git commit --message '",
                    CommitMessage,
                    "' \n git push origin master \n",
                    sep = ""
                    )

system(command = gitCommand,
       intern = TRUE
)

---

23 Her dökümanın sonuna kullandığınız kütüphaneler için atıf yazdırabilirsiniz

{r library citation, echo=TRUE}
citation()

---

23.1 Libraries Used

citation()

---

23.2 Bu oturuma spesifik kullanılan paketler

report::cite_packages(session = sessionInfo())

---

23.3 Tek tek paket atıfları

{r library citations}
citation("tidyverse")

{r}
citation("readxl")
citation("janitor")
citation("report")
citation("finalfit")
citation("ggstatplot")

---

23.4 Jamovi ve R için atıf örneği

The jamovi project (2019). jamovi. (Version 0.9) [Computer Software]. Retrieved from https://www.jamovi.org. R Core Team (2018). R: A Language and envionment for statistical computing. [Computer software]. Retrieved from https://cran.r-project.org/. Fox, J., & Weisberg, S. (2018). car: Companion to Applied Regression. [R package]. Retrieved from https://cran.r-project.org/package=car.

---

24 Her dökümanın sonuna oturum detaylarınızı yazdırabilirsiniz

{r session info, echo=TRUE}
sessionInfo()

---

24.1 Session Info

sessionInfo()

---

25 Sonraki Konular

• RStudio ile GitHub kullanımı
• …

---

26 Sunum Linkleri

https://sbalci.github.io/MyRCodesForDataAnalysis/R-Markdown.nb.html https://sbalci.github.io/MyRCodesForDataAnalysis/R-Markdown.html

https://forms.gle/UqGJBiAjB8uLPRon8

---

27 Geri Bildirim

• Geri bildirim için tıklayınız: Geri bildirim formu

---

---

28 İletişim

Completed on 2019-09-30.

Serdar Balci, MD, Pathologist
drserdarbalci@gmail.com

https://rpubs.com/sbalci/CV
https://sbalci.github.io/
https://github.com/sbalci
https://twitter.com/serdarbalci

---

---

29 Other Links

https://andrewbtran.github.io/NICAR/2018/workflow/docs/02-rmarkdown.html

• Troubleshooting in R Markdown

https://smithcollege-sds.github.io/sds-public/rmarkdown_problems.html

http://kbroman.org/knitr_knutshell/pages/Rmarkdown.html

https://kbroman.org/knitr_knutshell/pages/overview.html

https://kbroman.org/knitr_knutshell/pages/Rmarkdown.html

https://kbroman.org/knitr_knutshell/pages/markdown.html

https://onp4.com/

csv {headers: true, title: "**Drawing Tables In Markdown**"}
Name, Surname, Known As, Age
Marcelo, David, coldzera, 22
Oleksandr, Kostyliev, s1mple, 19
Nikola, Kovač, NiKo, 20
Richard, Papillon, shox, 25
Nicolai, Reedtz, dev1ce, 21

{pgn}
[Event "Bled-Zagreb-Belgrade Candidates"]
[Site "Bled, Zagreb & Belgrade YUG"]
[Date "1959.10.11"]
[Round "20"]
[Result "1-0"]
[White "Mikhail Tal"]
[Black "Robert James Fischer"]

1. d4 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5.
Be2 O-O 6. Nf3 e5 7. d5 Nbd7 8. Bg5 h6 9.
Bh4 a6 10. O-O Qe8 11. Nd2 Nh7 12. b4 Bf6
13. Bxf6 Nhxf6 14. Nb3 Qe7 15. Qd2 Kh7 16.
Qe3 Ng8 17. c5 f5 18. exf5 gxf5 19. f4 exf4
20. Qxf4 dxc5 21. Bd3 cxb4 22. Rae1 Qf6 23.
Re6 Qxc3 24. Bxf5+ Rxf5 25. Qxf5+ Kh8 26.
Rf3 Qb2 27. Re8 Nf6 28. Qxf6+ Qxf6 29. Rxf6
Kg7 30. Rff8 Ne7 31. Na5 h5 32. h4 Rb8 33.
Nc4 b5 34. Ne5 1-0

• Keeping Credentials Secret with Keyrings in R

https://ras44.github.io/blog/2019/01/19/keeping-credentials-secret-with-keyrings-in-r.html

• How to build a website with Blogdown in R

http://www.storybench.org/how-to-build-a-website-with-blogdown-in-r/

---

1. Bu bir derlemedir, mümkün mertebe alıntılara linklerle referans vermeye çalıştım.↩︎